There have been some recent studies documenting the changes in microsatellite polymorphisms in malignant cells in comparison with normal cells obtained from the same patient. The specific aim of this proposal is to repeat these studies in our institution with two goals: 1) to confirm that microsatellite amplification can indeed detect differences in the genomes of (i) malignant cells in urine specimens and (ii) normal cells, such as lymphocytes, from the same patient; and 2) to refine the methodology such that early premalignant cells can also be detected.